Comparing Traditional and Newer Definitions of Obesity in Relation to Incident Cardiovascular Disease and Obesity-Related Cancer Risk: A Prospective Cohort Study in ARIC

Authors:
Preprint
Publisher website
Google Scholar
Abstract
Background: Obesity is a risk factor for both cardiovascular disease (CVD) and cancer. The traditional body mass index (BMI)-based obesity definition (BMI ≥30 kg/m²) has limitations across adulthood and ancestry. The Global Commission on Clinical Obesity in 2025 proposed a new definition incorporating central adiposity measures. The magnitude of increased disease risk among the newly classified obese persons remains undetermined. We compared the associations between obesity definitions and incident CVD and obesity-related cancer in a community-based cohort. Methods: We analyzed 14,834 Atherosclerosis Risk in Communities (ARIC) study participants followed for a median of 25 years. Participants were classified as obese using the traditional definition (BMI ≥30 kg/m²) and the new definition (BMI ≥25 kg/m² plus ≥1 central adiposity measure, or two central adiposity measures as follows: waist circumference ≥102 cm Male/ ≥88 cm Female; waist-to-hip ratio >0.90 Male/ >0.85 Female; waist-to-height ratio >0.5). We identified participants classified as obese only by the new definition. Cox models adjusted for demographics, socioeconomic, and clinical risk factors estimated the adjusted hazard ratios (aHR). A subgroup analyses by age, sex, and race, and time-dependent sensitivity analyses were conducted. Results: 54% and 27% were classified as obese, respectively, according to the new-only (median BMI 26 kg/m 2 ) versus the traditional (median BMI 33 kg/m 2 ) definition; 19% were non-obese by both. 7% of the “obese by new-only definition” group (i.e. BMI 25-2 and central adiposity) reverted to non-obese status. Compared to non-obese, both new-only (aHR 1.25, 95% CI 1.15-1.35) and traditionally-defined (aHR 1.65, 95% CI 1.51-1.81) obesity were associated with higher CVD risk. The same pattern was noted in coronary heart disease and heart failure. For obesity-related cancers, the traditional definition (aHR 1.36, 95% CI 1.16-1.59), but not the new-only definition (aHR 1.09, 95% CI 0.95-1.26), conferred significantly higher risk. The traditional definition was also associated with higher risk of total and colorectal cancers. Time-dependent analysis and subgrouping by sex, race, and age at time of diagnosis yielded similar results. Conclusion: Both obesity definitions identified increased CVD risk, with significant trends across groups. Only traditionally-defined obesity consistently showed increased risk for total and obesity-related cancers. The newer, more sensitive, definition identified a large cohort with intermediate CVD risk, not previously captured by BMI alone, highlighting a distinct risk profile, suggesting potential for targeted interventions. These findings warrant careful consideration of potentially inflated risk in patients defined as obese by the new criteria. Funding: NHLBI, NCI, NPCR, AHA

This publication has 38 references indexed in Scilit: